The objective of the project is to map urban heat hotspots and identify the most suitable locations for cooling interventions such as tree planting or permeable surface expansion. Land Surface Temperature (LST) will be primarily derived from Landsat 8/9 thermal infrared data, complemented by additional sources such as Sentinel-3 SLSTR, ECOSTRESS and Copernicus Urban Atlas datasets. The forthcoming ESA LSTM (Land Surface Temperature Monitoring) mission can also be considered for future integration to enable higher spatio-temporal resolution. Optical indicators from Sentinel-2 and building footprint data shall be combined to classify surface types and evaluate greening potential. Recent scientific studies suggest that AI-based models can improve sub-pixel thermal downscaling and predict nocturnal heat retention, which could be assessed for incorporation into the methodology. The results shall be delivered via an interactive platform to support municipalities in climate adaptation planning.
Core components
- Thermal environment mapping: Land Surface Temperature (LST) will be derived primarily from Landsat 8/9 Thermal Infrared Sensor (TIRS) data, complemented by Sentinel‑3 SLSTR and NASA ECOSTRESS products to capture diurnal temperature dynamics. In preparation for higher-resolution thermal data, the upcoming ESA LSTM (Land Surface Temperature Monitoring) mission will be conceptually integrated for future scalability.
- Surface cover and greening potential analysis: Sentinel‑2 optical bands, vegetation indices (e.g., NDVI, NDBI, SAVI), and urban land-use layers from the Copernicus Urban Atlas and OpenStreetMap will be combined with building footprint datasets to distinguish between built-up areas, vegetation, water, and impervious surfaces. Potential cooling intervention zones (e.g., unsealed surfaces, rooftops suitable for greening, or low-canopy-density streets) will be identified through rule-based and machine learning classification approaches.
- AI-based modeling: Recent studies have demonstrated the ability of artificial intelligence to improve thermal downscaling, prediction of nocturnal heat retention, and identification of microclimatic anomalies. The project will explore suitable model architectures—such as convolutional neural networks (CNNs) or random forest regression—for producing high-resolution heat and greening suitability maps.
Expected research outcomes
- A multi-source workflow for integrating thermal, optical, and urban vector data within a unified spatial reference framework.
- Quantitative mapping of urban heat hotspots and mitigation potential for selected Hungarian cities.
- Comparison of traditional statistical vs. AI-based methods for LST downscaling and urban thermal modeling.
- Development of an interactive, scalable interface for data visualization and stakeholder engagement.
Implementation platforms
Students may start the project development in local or simplified computing environments (e.g., Python-based Jupyter notebooks using downloaded datasets) to prototype algorithms, test data preprocessing workflows, or train preliminary models on subsets of the data. This allows for flexible experimentation and debugging before scaling up.
However, the goal is to deliver a production-ready, scalable system capable of automated, continuous operation and ready for national deployment. For this purpose, a containerized server / cloud infrastructure or open data infrastructures can be leveraged. Students are especially encouraged to use Google Earth Engine (GEE) or the Copernicus Data Ecosystem to manage, process, and visualize large-scale datasets efficiently.
- Google Earth Engine offers cloud-based access to Sentinel data, integrated climate sources, and scalable machine learning tools, enabling rapid prototyping and visualization.
- Copernicus Data Space Ecosystem provides APIs and interactive tools for direct access to high-volume Sentinel archives and Copernicus service data, supporting reproducible scientific workflows and integration with AI frameworks (e.g., TensorFlow, PyTorch).
Students are encouraged to select a platform according to their experience and research focus—starting locally for development and transitioning toward an operational prototype hosted on GEE, Copernicus, or a dedicated server-based environment capable of real-time updates and user access.